- Top of page
- MATERIALS AND METHODS
To understand more clearly how mucosal and systemic immunity is regulated by ovarian steroid hormones during the menstrual cycle, we evaluated the frequency of immunoglobulin- and antibody-secreting cells (ISC, AbSC) in genital tract and systemic lymphoid tissues of normal cycling female rhesus macaques. The frequency of ISC and AbSC was significantly higher in tissues collected from animals in the periovulatory period of the menstrual cycle than in tissues collected from animals at other stages of the cycle. The observed changes were not due to changes in the relative frequency of lymphocyte subsets and B cells in tssues, as these did not change during the menstrual cycle. In vitro, progesterone had a dose-dependent inhibitory effect, and oestrogen had a dose-dependent stimulatory effect on the frequency of ISC in peripheral blood mononuclear cell (PBMC) cultures. The in vitro effect of progesterone and oestrogen on ISC frequency could not be produced by incubating enriched B cells alone with hormone, but required the presence of CD8+ T cells. Following oestrogen stimulation, a CD8+ enriched cell population expressed high levels of IFN-gamma and IL-12. The changes in B cell Ig secretory activity that we document in the tissues of female rhesus macaques during the menstrual cycle is due apparently to the action of ovarian steroid hormones on CD8+ T cells. Thus, CD8+ T cells control B cell secretory activity in both mucosal and systemic immune compartments. Understanding, and eventually manipulating, the CD8+ regulatory cell–B cell interactions in females may produce novel therapeutic approaches for autoimmune diseases and new vaccine strategies to prevent sexually transmitted diseases.
- Top of page
- MATERIALS AND METHODS
Sex steroid hormones are involved in regulating the immune system [1,2] and autoimmune diseases effect women to a greater degree than men (reviewed in ). Multiple sclerosis and rheumatoid arthritis occur between two and three times more often in women and systemic lupus erythematosus affects nine times more women than men (reviewed in ). Oestrogen is the defining ovarian steroid hormone and effects of oestrogen are mediated by two distinct intracellular receptors, ER-alpha and ER-beta . It seems that women have a greater susceptibility to autoimmune disease at least in part because they make stronger immune responses. The levels of IgM in a primary response, and IgG in a secondary immune response are higher in sexually competent female mice compared to male mice of equivalent age [5–7]. In humans, plasma IgM levels  and peripheral CD4+ T cell counts are higher in women than in men .
In addition to stronger immune responses and higher concentrations of antibodies and lymphocytes, regulation of immunity is complex in females because lymphocytes respond to changing concentrations of steroid sex hormones. There is a dramatic reduction in the number of activated and committed B cell precursors in the bone marrow of pregnant mice, and this is thought to be due to the elaboration of sex steroids during pregnancy [10,11]. In mice, long-term exposure to high doses of exogenous oestradiol enhances polyclonal B cell activation . In-vitro, oestrogen enhances non-specific differentiation of human immunoglobulin-secreting cells (ISCs) [13–15] and this oestrogen-mediated enhancement is due to inhibition of suppressor T cells . It should be clear from this very brief review that ovarian steroid hormones can regulate immune cell function.
Ovarian sex steroids levels have a particular influence on female genital tract immunity. In the rat, the stage of the oestrous cycle influences the accumulation of IgA and IgG in uterine secretions . IgA and IgG levels in cervical secretions of healthy women are lowest during the periovulatory stage of the menstrual cycle . Within the narrow window of the periovulatory period, IgA levels in human cervical mucus are maximal 2–3 days before ovulation and drop to their minimum level at ovulation . Thus, mucosal immune responses in the female genital tract, as measured by immunoglobulin (Ig) or antibody levels, vary during the menstrual cycle. We recently demonstrated that the effect of menstrual cycle stage on IgG and IgA levels in cervico-vaginal secretions of macaques is similar to women . Thus, in the cervico-vaginal secretions of both monkeys and women, IgA and IgG levels are highest in the luteal phase and during menstruation and lowest around ovulation. We also demonstrated that the changes in Ig levels are not due to changes in immune cell populations in the genital tract . Thus, it seems most likely that functional differences in B cell activity account for the observed cyclicity of Ig levels in female genital tract secretions.
However, no information is available regarding the role of sex steroids on B cell physiology of primates in vivo. In the present study, we document profound changes in the frequency of ISC and antibody-secreting cells (AbSC) in the mucosal and systemic lymphoid tissues of rhesus monkeys at different stages of the menstrual cycle. Further, progesterone suppresses ISC frequency in monkey PBMC in vitro and oestrogen enhances ISC frequency in vitro. The effect of progesterone and oestrogen on the frequency of ISC in vitro could not be elicited by hormone treatment of enriched-B cells, but required the presence of CD8+ T cells in the cultures. The indirect effect of ovarian steroids on B cell function in vitro is consistent with the observed, menstrual cycle-related, variations in the frequency of ISC and AbSC in lymphoid tissues of female rhesus macaques. The results of these studies provide the first clear link between the effects of ovarian hormones on B cell function in vitro and the relative number of antibody-secreting B cells isolated from tissues at different stages of the menstrual cycle
- Top of page
- MATERIALS AND METHODS
In the present study we demonstrate, for the first time, that the frequency of ISC and AbSC in the genital tract mucosa and systemic lymphoid tissues vary markedly at different stages of the menstrual cycle of primates. The frequency of IgG and IgA-ISC (specific and total) in the cervico-vaginal tissues was highest during the periovulatory stage and lowest during the luteal phase and menstruation. We also provide the first evidence that these changes in the frequency of ISC occur in lymphoid tissues throughout the body during the menstrual cycle. Further, the in-vivo results are consistent with the findings of in vitro experiments. In vitro, oestrogen consistently increased the frequency of ISC (both IgG and IgA), while progesterone decreased the frequency of ISC in rhesus monkey PBMC. The effect of the ovarian steroids on monkey B cells was dose-dependent and required the presence of CD8+ T cells in the cultures. This result is consistent with the published findings that oestrogens enhance non-specific differentiation of human antibody-secreting cells in vitro[14,15] by inhibition of suppressor T cells . As we cannot exclude the possibility that CD8+ NK cells contribute to the observed effects, experiments are under way to define further the CD8+ cell subset that is responsible for altering ISC frequency. However, in vitro studies in rhesus monkeys and humans support the hypothesis that CD 8+ T cells can regulate Ig secretion by B cells in response to changes in ovarian hormone levels. These observations are also the first direct experimental evidence for an in vivo role of ovarian sex steroids and CD8+ T cells on ISC frequency in lymphoid tissues of menstrual monkeys and, by extension, women.
While our in vivo findings are in agreement with the in vitro studies, the timing of peak ISC frequency at ovulation was surprising. Previous reports, including our own, have shown that the levels of IgG and IgA in cervico-vaginal secretions of humans and rhesus macaques are the lowest around ovulation [19,23,24]. There may be several explanations for the discordance between the relatively low Ig levels in cervico-vaginal secretions and the high frequency of ISC in the lower genital tract mucosa at mid-cycle. The most likely explanation is that the increase in local Ig production does not compensate for the dilution effect caused by the increase in the production of cervical mucus at mid-cycle [24–26]. In addition, the stratified squamous epithelium of the vagina and ectocervix of these animals, which is the primary barrier to diffusion of IgG and monomeric IgA from the lamina propria to vaginal secretions, is thickest at ovulation . Although the concentrations of IgG and IgA in serum are not influenced by the menstrual cycle , the frequency of ISC in numerous tissues was significantly higher at mid-cycle than at other stages of the cycle. Similar trends toward a higher frequency of ISC at mid-cycle, although not statistically significant, were seen in most other tissues examined. Reconciling the stable serum Ig levels with the observed changes in ISC frequency is straightforward given that the half-life of IgG and monomeric IgA are 22 and 6 days, respectively . The relatively long half-life of serum Ig would mask changes in Ig levels that might be caused by the cyclic variations in number of ISC.
The observed changes in ISC frequency could reflect changes in the relative frequency of other mononuclear cell populations in the tissues. However, quantitative morphometric analysis of mononuclear cell populations demonstrated that there were no significant differences in the size or composition of the lymphocyte, macrophage or dendritic cell populations within the lower genital tract tissues of the 18 animals in this study . In addition, there are no significant changes in the absolute number of lymphocytes or the relative frequencies of lymphocyte subsets in the peripheral blood of female rhesus macaques that could explain the observed shifts in ISC frequency during the menstrual cycle (Miller, unpublished data).
In cultures of rodent and human PBMC, oestrogen stimulates pokeweed mitogen-induced B cell differentiation [14,28]. The immunostimulatory effect was thought to have been mediated by the inhibitory effect of oestrogen on a T suppressor cell population . The results of the in-vitro experiments with monkey PBMC suggest that the immunostimulatory effect of oestrogen on B cells is CD8+ T cell-mediated. We also found that oestrogen treatment resulted in an up-regulation of IL-12, IFN-γ and MDC mRNA in CD8 positive cells. The up-regulation of IFN-γ mRNA suggests a CD8+ T cell-dependent, cytokine-mediated mechanism of oestrogen action. Consistent with this, it has been shown in that human CD8+ T cells, but not CD4+ T cells, express oestrogen receptors . In addition, in mice oestrogen directly increases the activity of the IFN-γ promoter in lymphoid cells . Our results are also in agreement with data reporting an increased secretion of IFN-γ in stimulated and unstimulated human PBMC after oestrogen treatment .
IL-12 is a cytokine produced by macrophages and dendritic cells (DC) [33–35]. There are no reports of IL-12 production by CD8+ T cells, thus it difficult to explain the increased IL-12 expression found in the oestrogen-stimulated CD8+-enriched cultures. The simplest explanation is that antigen-presenting cells (APCs) were present in the CD3–CD8– cells that comprised 15% of the CD8+ cell-enriched fraction or that CD8αα+ DC were present in the CD8+-enriched cultures. Indeed, based on these considerations, it seems very likely that oestrogen stimulates IL-12 secretion in antigen-presenting, possibly dendritic cells (DC). It has been shown that DC-derived IL-12 is necessary for the differentiation of activated naïve B cells into plasma cells  and it is known that macrophages express oestrogen receptors . Thus, the source of the MDC and IL-12 in the CD8+ enriched fraction of oestrogen-stimulated rhesus monkey PBMC cultures could include APCs. It remains to be determined if oestrogen effects the number of immunoglobulin-secreting B cells via cytokine secretion by regulatory CD8 positive T cells alone or if DC are also involved.
Although we did not determine the mechanisms by which hormones regulate ISC frequency, a plausible explanation is that oestrogen increases the frequency of ISC by stimulating regulatory CD8+ T cells to increase expression of IFN-γ as observed. Increased IFN-γ expression in turn results in the improved differentiation of B cells and/or increased production of Ig by differentiated B cells. APCs secreting IL-12 may contribute to this T cell/B cell interaction. By contrast, we did not detect altered cytokine expression in progesterone-stimulated CD8+ T cells, suggesting that these soluble mediators do not mediate the progesterone-mediated decrease in ISC frequency. In mice, there are several models of Ig suppression due to MHC I-restricted CD8+ T cell-mediated lysis of B cells [38–42]. By analogy, the progesterone-mediated decline in ISC cell frequency may be due to lysis of B cells by CD8+ T cells. Experiments to distinguish among these possibilities are under way.
The study also produced significant insight into the anatomical location of anamnestic B cell responses. The frequency of ISC found in the PBMC of the animals in this study is within the range described previously for humans and macaques. A comparison of the frequency of IgG-ISC in human and monkey PBMC as detected by ELISPOT assays developed in various laboratories, including ours, is provided in Table 2. Healthy humans  and rhesus macaques have a similar number of Ig-ISC in PBMC (Table 2). We did, however, find a very high number of anti-TT AbSC in the iliac lymph nodes of experimental animals (Fig. 3). This finding probably reflects the fact that TT is a recall antigen for all monkeys at the CRPRC, as they are immunized routinely with TT beginning at a young age. Thus, it is not surprising that the frequency of AbSC specific to the recall antibody, TT, was higher in the tissues tested than the frequency of AbSC producing antibodies against KLH, a novel Ag. The high frequency of anti-TT AbSC and anti-KLH AbSC in the iliac lymph nodes is consistent with the observation that multiple intramuscular immunizations induce the strongest immune response in local draining lymph nodes compared to other tissues .
Table 2. IgG-immunoglobulin secreting cell frequency in human and non-human primate PBMC
|Species||Immune status||n||IgG-ISC/106 MNC||Boost to sampling interval||References|
|Human||Normal||3||260 ± 74|| ||(49)|
|Human||Acute HIV infection||4||1500||<90d||(47)|
|Human||Chronic HIV infection||11||2500||4 months–12 years||(47)|
|Cynomologous monkeys||Normal||2||255 ± 115|| ||(43)|
|Cynomologous monkeys||Anamnestic (3×)||5||5850 ± 2650||7d||(43)|
|Rhesus monkeys||Normal||20||79 ± 18|| ||Miller, Lü (unpublished)|
|Rhesus monkeys||Immunized (3×)||5||344 ± 55||>21d||Miller, Lü (unpublished)|
|Rhesus monkeys||Anamnestic||17||8541 ± 2871||7d||See results|
The frequency of anti-TT IgG-AbSC in the cervico-vaginal mucosa and iliac lymph node did not vary significantly during the menstrual cycle, although the frequency of anti-TT IgG-AbSC in the cervico-vaginal mucosa was higher in Group II than in other groups. Hormone-induced variation in the anti-TT IgG-AbSC frequency may have been masked by the proliferation of anti-TT B cells during the strong and ongoing anamnestic response to the booster TT immunization. While anti-TT IgA-AbSC were undetectable in the cervico-vaginal mucosa, they were found in the iliac lymph nodes and the frequency of anti-TT IgA-AbSC was significantly higher in iliac lymph nodes of Group II animals compared to the other two groups.
In a separate and unrelated study, we previously reported  that oral CT immunization induced anti-CT AbSC formation predominantly in the mesenteric lymph nodes, but failed to induce ISC in the lower reproductive tract. The results of the current study further support that result, as anti-CT AbSC were found only in the mesenteric lymph nodes. It is important to note that in the current study, intramuscular (upper leg) immunization induced a high frequency of anti-TT IgG AbSC in the cervico-vaginal mucosa. We have also found that intranasal immunization of rhesus macaques induces antigen-specific B cell immunity in the lower genital tract . Thus it appears that immunization in some, but not all, mucosal or intramuscular sites can result in the localization of antigen-specific B cells in the cervico-vaginal mucosa of primates.
Although there have been numerous published in-vitro studies demonstrating that sex steroid hormones can effect human B-cell differentiation and function, there have been no reports establishing an in-vivo role for sex steroids in regulating B cell immunity of menstrual primates. In the current study, we show that, in rhesus monkeys, the frequency of ISC and AbSC in the genital tract tissues and numerous systemic lymphoid tissues are affected significantly by the stage of the menstrual cycle. Further, we found that the observed effect of oestrogen and progesterone on B cell physiology in-vitro is mediated indirectly through CD8+ T cells. Studies are under way to determine if a similar immune regulatory pathway exists in women. Autoimmune diseases are more common in women and many of these diseases are associated with a polyclonal activation of B cells or the production of self-reactive antibodies. In addition, it has proven to be very difficult to elicit protective immunity against sexually transmitted diseases in women. A better understanding of the hormone-mediated molecular and cellular regulatory pathways that control antibody secretion by B cells in females may lead to therapeutic interventions for immune-mediated diseases and vaccine strategies that are capable of protecting women, and their infants, from STDs.